CN-121993162-A - Deep coal seam collapse pressure prediction method based on logging parameter uncertainty analysis

CN121993162ACN 121993162 ACN121993162 ACN 121993162ACN-121993162-A

Abstract

The invention discloses a deep coal seam collapse pressure prediction method based on uncertainty analysis of logging parameters, and relates to the technical field of coal seam gas drilling engineering, the method comprises the steps of acquiring a section to be logged, and acquiring data of the section to be logged according to preset indexes based on preset measurement intervals to obtain a logging parameter sequence; the method comprises the steps of performing basic type uncertainty analysis and new type uncertainty analysis on a logging parameter sequence to obtain basic type uncertainty analysis results and new type uncertainty analysis results, integrating the basic type uncertainty analysis results and the new type uncertainty analysis results to obtain logging parameter uncertainty characteristics, and gradually predicting deep coal seam collapse pressure by combining the logging parameter uncertainty characteristics to obtain collapse pressure prediction results. The method solves the technical problem that the prediction result is not unique due to uncertainty of logging parameters in deep coal seam collapse pressure prediction, achieves the technical effects of avoiding interference of parameter fluctuation on the prediction result and improving reliability of the deep coal seam collapse pressure prediction result.

Inventors

CUI BIN
YANG SONG
LIU XIAO
SHAO XIAOPING
ZHAO JINGHUI
ZHANG GUANGRONG

Assignees

中国石油化工股份有限公司
中国石油化工股份有限公司华东油气分公司

Dates

Publication Date: 20260508
Application Date: 20260303

Claims (10)

1. The method for predicting the collapse pressure of the deep coal seam based on the uncertainty analysis of the logging parameters is characterized by comprising the following steps of: acquiring a section to be measured, and acquiring data of the section to be measured according to a preset index based on a preset measurement interval to obtain a well measurement parameter sequence; Performing basic type uncertainty analysis and new type uncertainty analysis on the logging parameter sequence to obtain basic type uncertainty analysis results and new type uncertainty analysis results; Integrating the basic type uncertainty analysis result and the new type uncertainty analysis result to obtain logging parameter uncertainty characteristics; And gradually predicting the collapse pressure of the deep coal seam by combining the uncertainty characteristic of the logging parameter to obtain a collapse pressure prediction result.
2. The method for predicting collapse pressure in a deep coal seam based on uncertainty analysis of well logging parameters of claim 1, wherein the predetermined indicators include density, sonic jet lag, and natural gamma logging data.
3. A method of deep coal seam collapse pressure prediction based on logging parameter uncertainty analysis as defined in claim 1, wherein performing a base class uncertainty analysis and a new class uncertainty analysis on the sequence of logging parameters to obtain a base class uncertainty analysis result and a new class uncertainty analysis result comprises: Disassembling the logging parameter sequence to obtain a density sequence, a sound wave time difference sequence and a natural gamma logging data sequence; performing base class uncertainty analysis on the density sequence to obtain a density base class uncertainty analysis result; performing new type uncertainty analysis on the density sequence to obtain a new type uncertainty analysis result of the density; respectively adding the density base class uncertainty analysis result and the density new class uncertainty analysis result into the base class uncertainty analysis result and the new class uncertainty analysis result; and performing basic type uncertainty analysis and new type uncertainty analysis on the acoustic time difference sequence and the natural gamma logging data sequence, and adding analysis results into the basic type uncertainty analysis results and the new type uncertainty analysis results.
4. A method of predicting deep coal seam collapse pressure based on log parameter uncertainty analysis as recited in claim 3, wherein performing a basis type uncertainty analysis on the density sequence to obtain a density basis type uncertainty analysis result comprises: Performing jump point and missing value denoising screening on the density sequence to obtain a screening density sequence; Equidistant resampling is carried out on the screening density sequence to obtain an updated density sequence; And carrying out probability distribution identification on the screening density sequence through a nuclear density curve to obtain a density base class uncertainty analysis result.
5. A method of predicting deep coal seam collapse pressure based on log parameter uncertainty analysis as recited in claim 3, wherein performing a new type of uncertainty analysis on the density sequence to obtain a new type of density uncertainty analysis result comprises: carrying out gradient trend segmentation on the density sequence to obtain a plurality of density subsequences; extracting a first density subsequence from the plurality of density subsequences to perform new class uncertainty analysis, and obtaining a first density new class uncertainty sub-analysis result; And adding the first density new class uncertainty sub-analysis result into a density new class uncertainty analysis result.
6. The method for predicting collapse pressure of a deep coal seam based on uncertainty analysis of well logging parameters of claim 5, wherein performing gradient trend segmentation on the density sequence to obtain a plurality of density subsequences comprises: performing adjacent gradient identification on the density sequence to obtain an adjacent gradient sequence; identifying adjacent gradients exceeding a preset gradient threshold in the adjacent gradient sequence to obtain a plurality of cutting points; And mapping and cutting the density sequence based on the plurality of cutting points to obtain a plurality of density subsequences.
7. The method for predicting collapse pressure of a deep coal seam based on uncertainty analysis of well logging parameters of claim 5, wherein extracting a first density subsequence from the plurality of density subsequences for new type uncertainty analysis, obtaining a first density new type uncertainty sub-analysis result, comprises: extracting the average value of the first density subsequence as a new type uncertainty analysis starting point; Based on the first density subsequence, constructing a new class uncertainty analysis starting point neighborhood of a new class uncertainty analysis starting point according to a preset neighborhood bandwidth; According to a preset neighborhood bandwidth, carrying out neighborhood edge bidirectional diffusion on the new class uncertainty analysis starting point neighborhood to obtain a first diffusion new class uncertainty analysis starting point neighborhood; Comparing the neighborhood data volume difference between the first diffusion new type uncertainty analysis starting point neighborhood and the new type uncertainty analysis starting point neighborhood, and continuing to perform neighborhood edge bidirectional diffusion on the first diffusion new type uncertainty analysis starting point neighborhood according to a preset neighborhood bandwidth when the neighborhood data volume difference is larger than a preset difference threshold value until the neighborhood data volume difference obtained by two adjacent diffusion is smaller than or equal to the preset difference threshold value, so as to obtain a target diffusion new type uncertainty analysis starting point neighborhood; And carrying out probability distribution recognition based on the target diffusion new class uncertainty analysis starting point neighborhood to obtain a first density new class uncertainty sub-analysis result.
8. The method for predicting deep coal seam collapse pressure based on uncertainty analysis of well logging parameters as recited in claim 1, wherein integrating the base class uncertainty analysis result with the new class uncertainty analysis result to obtain the uncertainty characteristics of the well logging parameters comprises: performing similarity recognition on the basic class uncertainty analysis result and the new class uncertainty analysis result, and constructing an integrated similarity matrix; Performing convolution interaction on the basic type uncertainty analysis result and the new type uncertainty analysis result based on the integrated similarity matrix to obtain an interaction basic type uncertainty analysis result and an interaction new type uncertainty analysis result; and weighting the interaction base type uncertainty analysis result and the interaction new type uncertainty analysis result according to preset weights to obtain logging parameter uncertainty characteristics.
9. The method for predicting the collapse pressure of the deep coal seam based on the uncertainty analysis of the logging parameters as claimed in claim 1, wherein the step of predicting the collapse pressure of the deep coal seam by combining the uncertainty characteristics of the logging parameters to obtain a predicted collapse pressure result comprises the following steps: Determining rock mechanical parameter uncertainty characteristics of the rock mechanical parameters by combining the rock mechanical parameter model based on the logging parameter uncertainty characteristics; Substituting the mean value and the standard deviation of the uncertainty characteristic of the logging parameter and the uncertainty characteristic of the rock mechanical parameter into a stratum pressure calculation model to obtain the uncertainty characteristic of the stratum pressure; Deriving uncertainty characteristics of horizontal ground stress based on uncertainty characteristics of logging parameters, uncertainty characteristics of rock mechanical parameters and uncertainty characteristics of formation pressure by combining a Huang model and Rosenbluthe improvement method, and obtaining a mean value and a standard deviation of the uncertainty characteristics; The rock mechanical parameter uncertainty characteristic, the stratum pressure uncertainty characteristic and the horizontal ground stress uncertainty characteristic are synthesized to obtain the mean value, standard deviation and probability distribution interval of collapse pressure; And taking the mean value, standard deviation and probability distribution interval of the collapse pressure as the collapse pressure prediction result.
10. A method of predicting the collapse pressure of a deep coal seam based on analysis of uncertainty of a logging parameter as defined in claim 9, wherein the rock mechanical parameter uncertainty characteristics include modulus of elasticity, poisson's ratio, cohesion and internal friction angle; Formation pressure uncertainty characteristics include the mean and standard deviation of formation pressure.

Description

Deep coal seam collapse pressure prediction method based on logging parameter uncertainty analysis Technical Field The invention relates to the technical field of coal bed methane drilling engineering, in particular to a deep coal bed collapse pressure prediction method based on logging parameter uncertainty analysis. Background In low-carbon coal exploitation, the stability of a well wall of deep coal seam drilling is a key for guaranteeing safe and efficient propulsion of exploitation, and collapse pressure prediction is a core link. The prior art mainly relies on logging data such as density, acoustic time difference and the like to predict collapse pressure, and the method is applied to conventional coal seam drilling. However, in a deep coal seam scene, uncertainty exists in logging parameters, such as a testing method, instrument differences, human factors and underground environment interference, and the uncertainty can be transmitted to calculation of rock mechanical parameters, ground stress and the like, so that a collapse pressure prediction result is not unique. The collapse pressure is regarded as a fixed value by the traditional method, the actual situation cannot be accurately reflected, and the requirements of stable and accurate evaluation and management and control of the deep coal seam drilling well wall in coal low-carbon exploitation are difficult to meet. Disclosure of Invention The method solves the technical problem that the prediction result is not unique due to uncertainty of logging parameters in deep coal seam collapse pressure prediction. Aiming at the technical problems, the application provides a deep coal seam collapse pressure prediction method based on logging parameter uncertainty analysis, which comprises the steps of obtaining a logging section, carrying out data acquisition on the logging section according to preset indexes based on preset measurement intervals to obtain a logging parameter sequence, carrying out basic type uncertainty analysis and new type uncertainty analysis on the logging parameter sequence to obtain basic type uncertainty analysis results and new type uncertainty analysis results, integrating the basic type uncertainty analysis results and the new type uncertainty analysis results to obtain logging parameter uncertainty characteristics, and carrying out deep coal seam collapse pressure prediction step by combining the logging parameter uncertainty characteristics to obtain collapse pressure prediction results. The application provides one or more technical schemes, which at least have the following technical effects: According to the method, related parameter data are collected at preset intervals in a well section to be tested, the uncertainty characteristics of the parameters are obtained through basic type and new type uncertainty analysis and result integration, the uncertainty characteristics of rock mechanical parameters, formation pressure and horizontal ground stress are gradually deduced, and the mean value, standard deviation and probability distribution interval of collapse pressure are calculated by combining related criteria and an improvement method, so that the probability distribution condition of collapse pressure of a deep coal seam is accurately obtained, the prediction result of collapse pressure of the deep coal seam is more accurate and reliable, scientific support is provided for drilling engineering, interference of parameter fluctuation on the prediction result is avoided, and the technical effect of reliability of the prediction result of collapse pressure of the deep coal seam is improved. Drawings In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. Fig. 1 is a schematic flow chart of a deep coal seam collapse pressure prediction method based on uncertainty analysis of logging parameters according to an embodiment of the present application. Fig. 2 is a schematic flow chart of obtaining uncertainty characteristics of logging parameters in a deep coal seam collapse pressure prediction method based on uncertainty analysis of logging parameters according to an embodiment of the present application. Detailed Description The application provides a method for predicting the collapse pressure of a deep coal seam based on logging parameter uncertainty analysis, which solves the technical problem that the prediction result is not unique due to the logging parameter uncertainty in deep coal seam collapse pressure prediction. The technical solutions in the embodiments of the present application will be clearly and completely described below with referenc